14 and 18? (from this ask game https://www.tumblr.com/sesamie/804099166334828544)
14. If you could give a TED talk on anything, what would you talk about?
I’d talk about the work I’m doing on my workingline GSD database.
*climbs onto tiny soapbox*
I think it’s ridiculous that in 2025 we’re still making breeding decisions based on a static genealogy image (or a pedigree). In my breed, we pay the SV to rate our dogs’ hip, elbow, and spinal x-rays, and then they paywall that information. The SV database is very expensive, and the views are chopped up so you can’t even see all the information without a significant ongoing financial investment. If you try to cancel your subscription, then you are sent to an aggressive collection agency. Ask me how I know this. XD
Even our sources of pedigree information are either paywalled (workingdog.eu), buried under a wall of stroke-inducing ads (Pedigree Database), or contain sketchy information. While I am here ranting, why is it so hard to find out how a dog died? Why do people lie? If your GSD drops dead at six years old, I would put $1,000 down that the dog died of GDV, not "swallowed a ball".
I work in data science, and it makes me grind my teeth that we’re not taking advantage of basic visual reasoning and simple code to make connections that actually help people understand risk, health, and family lines. We chastise people for “not doing their research”, but how exactly are they supposed to do that? I’m competent at reading pedigrees, but it’s taken over a decade of consistent practice to build that skill. I was fortunate to learn at a time when experienced people were still willing to teach.
It’s been a fun project to work on in my free time. I think about how I read pedigrees and ask, "What feature would make this easier for me?". Then I write code to do that.
Here's the current version of the search interface:
Even simple features like saving a dog for export or marking a dog as seen are helpful.
I could go on, but I'll stop here. XD